Pelvic Floor Exercise for Urinary Incontinence

G Model MAT-5428; No. of Pages 7 ARTICLE IN PRESS

Maturitas xxx (2010) xxx–xxx

Contents lists available at ScienceDirect

Maturitas

journal homepage: www.elsevier.com/locate/maturitas

Review Pelvic ﬂoor exercise for urinary incontinence: A systematic literature review

Natalia Price ∗, Rehana Dawood, Simon R. Jackson

Department of Obstetrics and Gynaecology, John Radcliffe Hospital, Oxford OX3 9DU, UK article info abstract

Article history: Urinary incontinence is a common problem among adults and conservative management is recom- Received 10 July 2010 mended as the first-line treatment. Physical therapies, particularly pelvic floor muscle exercise, are the Received in revised form 6 August 2010 mainstay of such conservative management. The purpose of this review is to summarise current literature Accepted 10 August 2010 and describe trends in the use of pelvic floor muscle exercise in the management of urinary incontinence Available online xxx in women. Our review confirms that pelvic floor muscle exercise is particularly beneficial in the treatment of Keywords: urinary stress incontinence in females. Studies have shown up to 70% improvement in symptoms of Pelvic floor exercise Pelvic floor muscle training stress incontinence following appropriately performed pelvic floor exercise. This improvement is evident Urinary incontinence across all age groups. There is evidence that women perform better with exercise regimes supervised by Urinary stress incontinence specialist physiotherapists or continence nurses, as opposed to unsupervised or leaflet-based care. Overactive bladder syndrome There is evidence for the widespread recommendation that pelvic floor muscle exercise helps women Urge incontinence with all types of urinary incontinence. However, the treatment is most beneficial in women with stress urinary incontinence alone, and who participate in a supervised pelvic floor muscle training programme for at least three months. © 2010 Elsevier Ireland Ltd. All rights reserved.

Contents

1. Introduction ...... 00 2. Methodology ...... 00 2.1. Data sources ...... 00 2.2. Study selection ...... 00 2.3. Data extraction and quality assessment ...... 00 3. A short history of pelvic floor exercise ...... 00 4. Definition, prevalence and aetiology of urinary incontinence ...... 00 5. Pelvic floor exercise ...... 00 5.1. Muscle groups used in pelvic floor exercise ...... 00 5.2. Pelvic floor exercise regimens...... 00 6. How pelvic floor exercise works in treating stress urinary continence ...... 00 6.1. Strength training ...... 00 6.2. ‘The Knack’ manoeuvre ...... 00 6.3. Indirect training of pelvic floor muscles by contracting the abdominal muscles ...... 00 7. Biofeedback and other physical therapies ...... 00 7.1. Biofeedback therapy ...... 00 7.2. Vaginal cones ...... 00 7.3. Sacral nerve stimulation...... 00 7.4. Posterior tibial nerve stimulation ...... 00 7.5. Magnetic therapy ...... 00 9. Effectiveness of pelvic floor exercise in treatment of urinary stress incontinence ...... 00 9.1. Role of pelvic floor exercise in treatment of overactive bladder symptoms ...... 00 10. Pelvic floor exercise in pregnancy and postpartum ...... 00 11. Conclusions ...... 00

∗ Corresponding author. Tel.: +44 1865 221626; fax: +44 1865 221188. E-mail address: [email protected] (N. Price).

Contributors ...... 00 Competing interests ...... 00 Provenance ...... 00 References ...... 00

1. Introduction scheme, adequacy of randomisation and allocation concealment, and justification of sample sizes. Several strategies were used to Urinary incontinence (UI) is a common problem among adults reduce bias, including a comprehensive literature search of pub- living in the community. Its incidence increases with age and it lished and unpublished evidence in several databases, a search is more frequent in women, being particularly common amongst of reference lists of systematic reviews and proceedings of the elderly women in residential care. Estimates of the prevalence of International Continence Society, and contacts with experts for urinary incontinence in women vary from 10% up to 40% [1,2]. How- additional references. The quality of the selected studies was ever, these figures probably do not reflect the true scope of the assessed using a standard grading system, as outlined in the PRISMA problem, because of under-reporting arising from social embar- (Preferred Reporting Items for Systematic Reviews) statement. The rassment associated with the condition. studies were then combined quantitatively. Commercially avail- Pelvic floor exercise offers a possible reprieve from urinary able software was used for the analysis. incontinence [3]. This conservative therapy appears to have no significant side effects and enables improvement in symptoms; it can therefore be considered as a first choice of treatment for urinary incontinence in women. Moreover, if the outcome is unsatisfac- 3. A short history of pelvic floor exercise tory the patient can be referred for further evaluation and possible surgical intervention. The National Institute of Clinical Excellence Pelvic floor muscle training (PFMT) for the management of uri- guideline No. 40 on the management of urinary incontinence in nary incontinence has been described in several ancient texts. women recommends pelvic floor muscle training for at least three So-called “Deer Exercises” were part of the exercise routine in Chi- months as the primary treatment for urinary stress incontinence. nese Taoism for over 6000 years. Ancient Indian texts report similar The guideline states that pelvic floor exercises were found to be exercises as part of the Ashwini Mudra (“horse gesture”), practiced effective in the treatment of incontinence in female patients in by the Yogis. Hippocrates and Galen also described pelvic floor exer- more than 50% of cases [4]. cise regimens in the baths and gymnasiums of ancient Greece and The purpose of this review is to summarise recently published Rome [5]. It was thought that strengthening this group of muscles data on the use of pelvic floor muscle training for treatment for would promote health, longevity, spiritual development and sexual urinary incontinence. health. Pelvic floor muscle training first entered modern medicine in 2. Methodology 1936, when a paper by Margaret Morris, describing tensing and relaxing of the pelvic floor muscle as a preventative and treatment 2.1. Data sources option for urinary and faecal incontinence, introduced PFMT to the British physiotherapy profession. However, the use of PFMT In conducting this systematic review, we searched the MEDLINE as a treatment for stress urinary incontinence did not become (via PubMed), CINAHL and Cochrane databases for relevant articles widespread until after 1948, when Arthur Kegel, a professor of and undertook manual searches of reference lists from systematic obstetrics and gynaecology in the USA, established its regular prac- reviews and proceedings of the International Continence Society. tice. In his paper ‘Progressive resistance exercise in the functional restoration of the perineal muscles’ he reported the successful treat- 2.2. Study selection ment of 64 patients with urinary stress incontinence [6], hence the term Kegel Exercise, a common misnomer for pelvic floor exercises When deciding on study eligibility we followed the recom- as described by Kegel. mendations of the Cochrane Handbook for Systematic Reviews of Interventions and included original publications of randomised controlled trials (RCTs) that were published in English from 1990 to May 2010. Full texts of the RCTs that examined the effects of non-surgical clinical interventions on urinary incontinence in community-dwelling women were eligible for the review. We excluded secondary data analyses, case reports, case series and RCTs that did not report patient outcomes. We also excluded RCTs that analysed surrogate outcomes of subjective and objective measures of severity of urinary incontinence, including continuous changes in the number of incontinence episodes or pad use, and urodynamic variables (Fig. 1).

2.3. Data extraction and quality assessment

We extracted the data using standardised forms that elicited information about study samples, interventions, designs, and outcomes. Study quality was analysed using the following criteria: participant selection, length and loss of follow-up, use of intention- to-treat principle, masking of the treatment status, randomisation Fig. 1. Selection of studies for analysis.

4. Deﬁnition, prevalence and aetiology of urinary Table 1 incontinence Modiﬁed Oxford Pelvic Tone Scoring System. Oxford Score Signs

Urinary incontinence, as defined by The International Conti- 0 No contraction nence Society, is the complaint of any involuntary leakage of urine 1 A flicker [7]. It can result from a variety of different conditions and it is use- 2 Weak ful to classify them accordingly. The most common types of urinary 3 Moderate with some lift 4 Good contraction with lift, against some resistance incontinence in women are stress and urge incontinence. 5 Normal muscle contraction, strong squeeze and lift Urinary stress incontinence is the complaint of involuntary leakage of urine on effort or exertion, such as sneezing or coughing [7]. When urodynamic studies demonstrate the involuntary loss of ani muscles and the endopelvic fascia by way of stretching or tear- urine during increased intra-abdominal pressure not caused by a ing. It has been observed that bladder neck mobility is worsened contraction of the detrusor muscle, this is defined as urodynamic following vaginal births and this is postulated to be the cause of uri- stress incontinence. The involuntary leakage of urine, accompanied nary stress incontinence secondary to parturition. A study by van by or immediately preceded by a strong desire to pass urine (void), Brummen et al. showed that the antenatal development of stress is described as urge incontinence. urgency, with or without urge incontinence lead to an 18-times higher risk of developing stress urinary incontinence and usually with frequency and nocturia, is incontinence during the year following child birth, and that this also defined as overactive bladder syndrome (OAB) [7]. Mixed uri- was most prevalent in the group that delivered vaginally [13]. nary incontinence is when women have symptoms of both types of Other risk factors associated with urinary stress incontinence incontinence. Usually, one of these is predominant; that is, either are obesity (body mass index of over 30), high impact sports (e.g. the symptoms of urge incontinence, or those of stress incontinence, trampolining, pole vaulting), chronic respiratory disorders causing are most bothersome. a chronic cough, and intra-abdominal masses causing an increase Estimates of the prevalence of urinary incontinence in women in the intra-abdominal pressure. Medication such as diuretics, vary between 10% and 40% of the female population [1]. Factors endocrine disorders (diabetes), central or peripheral neuropathies, commonly found to affect the prevalence of urinary incontinence dementia and smoking are also well known culprits of urinary are: age, gender, race and residing in a nursing home. The preva- incontinence. lence of urinary incontinence has been reported to increase with A substantial proportion of patients with urinary incontinence age. Data from a large epidemiological study (27,936 Norwegian are postmenopausal. Evidently, a hypo-oestrogenic state in a women) suggest a gradual increase in prevalence with age to an woman is associated with thinning of the urethral mucosa, reduc- early peak at around mid life (mid 50s), followed by a slight decline tion in urethral closure pressure from loss of sphincter tone, and or stabilisation until about 70 years of age, when the prevalence alteration of the urethro-vesical angle. Fantl et al. performed a begins to rise steadily [8]. The relationship of urinary incontinence meta-analysis on the use of topical oestrogen for urinary incon- to age was investigated by both Rud et al. and Enhorning et al., who tinence, which supported the use of topical oestrogen therapy for found that maximum urethral closure pressures tend to decrease the management of urinary incontinence in women [14,15]. Also, with age. This decrease was found to be significant after the age the latest Cochrane review on the use of oestrogen therapy for of 36 years and they reported a 2–4% decrease in the functioning urinary incontinence in postmenopausal women concluded that of the urethra after the age of 40 years [9,10]. The second peak in topical oestrogen treatment for incontinence may improve or cure the incidence of urinary incontinence after 70 years of age can be it; however, there was little evidence from the trials on the period explained by an increase in urgency and urge incontinence, possibly after oestrogen treatment had finished and none about long-term due to low levels of oestrogen. effects [16]. There is a racial difference in the prevalence of urinary stress incontinence, which may be explained by differences in the bulk 5. Pelvic floor exercise of urethral muscle in different races. Afro-Carribeans, who are thought to have a low prevalence of urinary stress incontinence, 5.1. Muscle groups used in pelvic floor exercise were found to have greater urethral sphincter capacity, as evi- denced by higher density of urethral striated muscle fibres and The pelvic floor consists of a group of 12 striated muscles higher urethral closure pressures both during pelvic contraction arranged in 3 layers. This muscular plate expands from the pubic and at rest. Women of Afro-Carribean descent also have a larger lev- symphysis to the side walls of the ileum towards the coccyx. The ator ani cross-sectional area and muscle strength. This anatomical striated muscle fibres of each muscle run in the same direction in difference may explain the reduced prevalence of urinary inconti- each muscle but in a different direction to the other muscles of nence in this population [11]. the pelvic floor group. However, when the pelvic floor contracts, Life events having major implications for urinary incontinence it is always en masse, moving the pelvic girdle in one direction are pregnancy, child birth and menopause. Pregnancy and vaginal [17]. The only known voluntary function of the pelvic floor muscle delivery are considered to be the main risk factors for the devel- group is a mass contraction, best described as an inward lift and opment of urinary incontinence. It seems that the prevalence of squeeze around the urethra, vagina and rectum [18]. The function urinary incontinence increases during pregnancy and decreases of the pelvic floor muscles is to lend structural support to the pelvic following delivery, although postpartum prevalence still remains structures, the urethra, vagina and rectum. The Oxford Pelvic Tone higher than before pregnancy. Estimates of the prevalence of stress Scoring System is a commonly used method for assessing pelvic urinary incontinence during pregnancy vary between 6% and 67%, floor muscle tone (Table 1) [19]. and from 3% to 38% two to three months after delivery. Urinary incontinence increases with parity and, in primiparas who deliver 5.2. Pelvic floor exercise regimens vaginally, it has been associated with decreases in pelvic muscle strength of 22–35% between pregnancy and the postpartum period Pelvic floor muscle training involves the repetitive contraction [12]. of the pelvic floor muscle, which builds strength and perineal sup- Pregnancy and vaginal delivery are known to be associated with port, and improves muscle tone. As the pelvic floor is entirely damage to the pelvic floor innervation, direct trauma to the levator composed of striated muscle, the principles of strength training for

Please cite this article in press as: Price N, et al. Pelvic floor exercise for urinary incontinence: A systematic literature review. Maturitas (2010), doi:10.1016/j.maturitas.2010.08.004 G Model MAT-5428; No. of Pages 7 ARTICLE IN PRESS 4 N. Price et al. / Maturitas xxx (2010) xxx–xxx striated muscle should be followed when attempting to tone and 6.2. ‘The Knack’ manoeuvre strengthen the pelvic floor. The movement is a voluntary inward and upward contraction or squeeze of the pelvic floor. The number The term ‘The Knack’ was coined by Ashton-Miller in the orig- of contractions recommended across studies ranges from 8 to 12 inal study, because the simple English word ‘knack’ implies an contractions three times a day, to 20 contractions four times a day, adroit way of doing something. This manoeuvre is performed by to as many as 200 contractions per day. However, Arthur Kegel, consciously contracting the pelvic floor muscle prior to a physical the founder of contemporary pelvic floor exercises, in his 1948 stress and then maintaining the contraction during the stress. This paper recommended up to 500 contractions a day. The duration prevents the urethra and bladder base descending and enhances of ‘squeeze and hold’, or contraction, varies in published stud- continence. An intentional, effective pelvic floor muscle contraction ies from 4 s to 30–40 s [20,21]. The recommended posture to be (lifting the pelvic floor muscle in a cranial and forward direc- adopted during the prescribed exercise regimen also varies and tion) prior to and during effort or exertion clamps the urethra and includes sitting, kneeling, standing, lying down and standing with increases the urethral pressure, thereby preventing urine leakage legs astride. The recommended duration of the prescribed regimen (DeLancey et al.) [25]. Ultrasonography and magnetic resonance varies widely, from one week to six months, with three months imaging studies have demonstrated the cranial and forward move- being most frequently recommended. The National Institute for ment of the pelvic floor muscle during active contraction and the Clinical Excellence recommends a trial of supervised pelvic floor resulting impact on the urethral position, which supports this ratio- exercises, consisting of at least eight contractions three times a nale. day for a minimum of three months, as a first-line treatment for It would seem to be common sense that, if one contracts the urinary incontinence [4]. The International Consultation on Inconti- urethral and levator ani striated muscles just before and during the nence Committee recommends that supervised pelvic floor muscle moment of a stressor event, one can prevent urine loss. Unfortu- training for women with stress incontinence is maintained for 8–12 nately, many women seem unable to discover this ‘hidden’ self-care weeks before reassessment and possible referral for further man- mechanism on their own and need to be taught ‘The Knack’. Thus it agement if the patient has not improved sufficiently [22]. is plausible that part of the mechanism by which pelvic floor exer- The ‘Quick Flick’ is a technique for use by women with urge cises become effective in treating stress urinary incontinence could incontinence or mixed urinary incontinence. This exercise involves be an increased awareness and skill in learning to time the contrac- taking slow deep breaths, while contracting the pelvic floor muscles tion with the event that causes the leakage. Miller et al. showed that rapidly 3–5 times, when the urge to void is felt. This has been found this simple manoeuvre can reduce urinary leakage by 98.2% with a to suppress the urge to void. medium cough, and 73.3% with a deep cough, after only one week There is evidence suggesting that it may not be necessary to of training [26]. maintain a lifelong regime of pelvic floor exercise, although this may be desired. An optimal pelvic floor exercise regime would 6.3. Indirect training of pelvic floor muscles by contracting the change the morphology and position of the muscles to enable sub- abdominal muscles conscious contraction, a mechanism thought to occur in continent women. In addition, as with strength training of skeletal muscle, Pelvic floor muscle may be activated together with the abdom- less effort would be needed to maintain muscle tone than to build inal muscle. An increasing body of evidence suggests that active muscle mass initially [17]. contraction of the transversus abdominus muscle is associated with co-activation of the pelvic floor muscle. This has been demonstrated by ultrasound, electromyography and magnetic resonance 6. How pelvic floor exercise works in treating stress urinary imaging studies. However, contraction of the transversus abdomi- continence nus muscle does not appear to elevate the pelvic floor muscle all women and, when it does, it does not appear to be as effective as The objective of pelvic floor muscle exercise is to improve the a direct contraction of the pelvic floor muscle itself. Recent stud- timing of contractions, the strength of the pelvic floor muscles ies suggest that the relationship between pelvic floor muscle and and the stiffness of the pelvic floor muscles. The mechanisms of transversus abdominus muscle differs between continent and incon- action of pelvic floor exercises are threefold: strength training, tinent women, with the pelvic floor muscle being displaced less counterbalancing, and indirect training of the pelvic floor muscle during a transversus abdominus muscle contraction in women with by contracting the transverse abdominal muscle. stress urinary incontinence as compared to continent women. More research is needed to understand the effect of incontinence on reha- bilitation of the interaction between the transversus abdominus and 6.1. Strength training pelvic floor musculature in the treatment of urinary incontinence [16]. The bladder neck is supported by the pelvic floor muscles, which limit the downward movement of the urethra during exertion and thereby prevent leakage of urine (Bo 2004, Peschers 2001) [17,21]. 7. Biofeedback and other physical therapies Intensive training of any striated muscle will build muscle bulk; similarly strength training of the pelvic floor muscles will build Other physical therapies recommended for treatment of stress muscle bulk and thereby provide structural support to the pelvic urinary incontinence include biofeedback, the use of vaginal cones, floor by permanently elevating the levator muscle plate to a higher electrical stimulation, transcutaneous electrical nerve stimulation position in the pelvis. The support is further enhanced hypertro- and posterior tibial nerve stimulation, and magnetic therapy. phy and stiffness of the endopelvic fascia. Balmforth et al. reported increased urethral stability at rest and during effort following 14 7.1. Biofeedback therapy weeks of supervised pelvic floor muscle training. Pelvic floor muscle training will, in addition, facilitate more effective automatic motor Biofeedback therapy provides awareness of the physiological unit firing of the pelvic floor musculature, preventing pelvic floor action of the pelvic floor muscles by visual, tactile or auditory descent during increased intra-abdominal pressure, and hence pre- means. Vaginal cones attached to electrodes, manometry and elec- vent leakage of urine [17,23,24]. tromyography are examples of such means.

7.2. Vaginal cones or together with, pelvic floor muscle training [16,27,32]. There- fore, it is recommended that pelvic floor electrical stimulation and Weighted cones in the vagina can be used for strength training biofeedback should not be used as a routine part of pelvic floor mus- of the pelvic floor muscles. They are of varying weight and are used cle training. While there is no evidence of effectiveness for either typically for around 20 min a day, starting with the lower weight biofeedback or electrical stimulation, the information and support and progressing according to the individual’s ability to hold the generated by biofeedback may assist motivation for some women, cones. The cones can weigh from 20 to 150 g. For example, Mabella and electrical stimulation may be of value for those who are unable cones (Vitacon AS), weighing 20, 40 and 70 g each, were used in the to initiate a pelvic floor muscle contraction. Therefore, electrical randomised controlled trial by Bo et al. [27]. stimulation and biofeedback could be considered in women who cannot actively contract pelvic floor muscles, in order to aid moti- 7.3. Sacral nerve stimulation vation and adherence to therapy [4,20].Supervision of pelvic floor exercise Electrical stimulation of the sacral reflex pathway can be used to The evidence suggests that women do better in exercise regimes inhibit the reflex behaviour of the bladder. Nerve stimulation can that are supervised by specialist physiotherapists or specialist con- be achieved using surface electrodes, or transcutaneous needles, or tinence nurses, in contrast to unsupervised or leaflet assisted care. electrodes implanted close to nerves. Initially an electrode is placed The reasons for this may be that, in addition to exercise, such spe- via the sacral foramen alongside a sacral nerve (usually S3). In an cialists are likely to cover numerous other areas which may also alternative procedure, the electrode is connected by wires under impact on the pelvic floor, such as occupation, respiratory sta- the skin to an implanted programmable pulse generator that pro- tus, lifestyle issues, overall muscle fitness, diet and general health. vides stimulation within set parameters [28]. This technology has There is also likely to be better compliance with exercise regimes in been used for patients with overactive bladders, urgency incon- the long term, if each woman fully understands how she can help tinence, and voiding (retention of urine) difficulties, and for some herself and if she has had adequate time to address the problem patients with defecation problems. It has also been used in the man- with professional support. agement of chronic pelvic pain, although this is outside the scope of Slack et al. recommend a dedicated pelvic floor physiotherapy this review. In randomised trials about 50% of patients in the stim- service and have found a reduction of 33% in the surgical and uro- ulation group achieved complete continence or an improvement dynamic work load following the use of this service [31]. In their greater than 90% in the main incontinence symptoms, while a 50% study they used a pelvic tone of less than or equal to 3 as a crite- improvement in the main incontinence symptoms was observed in rion for recommending pelvic floor exercises for the treatment of about 87% [28]. urinary incontinence. Ishiko et al. advised a supplement of intrav- Sacral nerve stimulation (SNS) is recommended for the treat- aginal oestriol in postmenopausal women undertaking pelvic floor ment of urinary incontinence due to detrusor overactivity in exercise and found that this resulted in a higher cure rate of incon- women who have not responded to conservative treatments. tinence [33]. Women should be offered sacral nerve stimulation on the basis of their response to preliminary percutaneous nerve evaluation. 9. Effectiveness of pelvic floor exercise in treatment of Lifelong follow-up is recommended [4]. urinary stress incontinence 7.4. Posterior tibial nerve stimulation Daily pelvic floor muscle training is an effective treatment for stress or mixed urinary incontinence, compared with no treatment, Stimulation of the posterior tibial nerve (PTNS) delivers retro- over the short term. Other than occasional cases of pain or dis- grade stimulation to the sacral nerve plexus. The posterior tibial comfort, no other adverse effects have been noted. This evidence nerve contains mixed sensory motor nerve fibres that originate is derived from several large randomised controlled trials and two from the same spinal segments as the innervations to the bladder systematic reviews published in the Cochrane library [4,20]. and pelvic floor. The exact mechanism of action of neuromodu- A study by Cammu et al., comprising a 10-year follow-up of lation is unclear. The potential benefit of percutaneous posterior women after pelvic floor muscle exercise for stress incontinence, tibial nerve stimulation is that it may achieve the same neuromod- concluded that when pelvic floor muscle training is initially suc- ulatory effect as sacral nerve stimulation through a less invasive cessful there is a 66% chance that the favourable results will persist route. for at least 10 years [34]. In a randomised controlled trial of 100 patients comparing PTNS The trials suggest that the treatment effect (especially self with medication, 80% (35/44) of patients in the PTNS group and reported cure/improvement) might be greater in women with 55% (23/42) of patients in the medication group considered them- stress urinary incontinence participating in a supervised PFMT pro- selves to be cured or improved (p = 0.01). Both groups showed a gramme for at least three months [31]. It also seems that the similar statistically significant decrease in the number of voids effectiveness of PFMT does not decrease with age: in trials with per day, nocturia, urge incontinence and the number of moderate stress urinary incontinent older women it appeared that results for to severe urgency episodes per day. Quality of life was also sig- both primary and secondary outcome measures were comparable nificantly improved in both groups immediately after treatment to those in trials with younger women. [29,30].

7.5. Magnetic therapy 9.1. Role of pelvic floor exercise in treatment of overactive bladder symptoms Magnetic therapy aims to stimulate the pelvic floor muscles and/or sacral roots by placing them within an electromagnetic field. Pelvic floor muscle exercise can also be used in the management The women remain fully clothed throughout the procedure and of urgency and urge incontinence. The biological rationale is based may find the process more acceptable when compared with elec- on Godec’s observation that a detrusor muscle contraction can be trical stimulation [31]. inhibited by a pelvic floor muscle contraction induced by electrical At present there is no robust evidence that such additional stimulation [35]. Further, de Groat demonstrated that during urine physical therapies are any more successful when used instead of, storage there is an increased pudendal nerve outflow response to

Please cite this article in press as: Price N, et al. Pelvic floor exercise for urinary incontinence: A systematic literature review. Maturitas (2010), doi:10.1016/j.maturitas.2010.08.004 G Model MAT-5428; No. of Pages 7 ARTICLE IN PRESS 6 N. Price et al. / Maturitas xxx (2010) xxx–xxx the external urethral sphincter, increasing intraurethral pressure surgical procedure, and has fewer and milder side effects compared and representing what he termed a ‘guarding reflex’ for conti- to pharmacological treatment. nence [36]. Additionally, Morrison demonstrated that Barrington’s The treatment effect is usually enhanced when the PFMT pro- micturition centre excitatory loop switches on when bladder pres- gramme is taught and supervised by a specialist physiotherapist or sures are between 5 and 25 mm Hg, while the inhibitory loop specialist continence nurse. Additional physical therapies, such as is predominantly active above 25 mm Hg. Inhibition involves an electrical stimulation and biofeedback, are not recommended for automatic (unconscious) increase in tone for both the pelvic floor routine use during pelvic floor muscle training. However, they can muscle and the urethral striated muscle [37]. Thus, voluntary pelvic be considered in women who cannot actively contract their pelvic floor muscle contractions may be used to control urgency and urge floor muscles, in order to aid motivation and adherence to therapy. incontinence. After inhibiting the urgency to void and the detrusor In common with its use in older women with stress incon- contraction, the patient can reach the toilet in time to avoid urine tinence, there is evidence for the widespread recommendation leakage. However, the number, duration, intensity and timing of that PFMT is an appropriate treatment for women with persis- the pelvic floor muscle contraction required to inhibit a detrusor tent postpartum urinary incontinence. It is possible that the effects muscle contraction are not known. of PFMT might be greater with targeted rather than population- based approaches and in certain groups of women (for example: primiparous women or women who have had bladder neck hyper- 10. Pelvic floor exercise in pregnancy and postpartum mobility in early pregnancy, a large baby, or a forceps delivery). The limited nature of follow-up beyond the end of treatment in There is strong evidence to suggest that women, who do inten- the majority of the published studies means that the long-term out- sive supervised pelvic floor exercises during pregnancy, reduce comes of PFMT are less clear. Longer-term effects may be greater in their chances of leakage postpartum in the first year after childbirth. women participating in supervised PFMT for at least three months. For women having their first baby, antenatal pelvic floor exercise Continued adherence to training may be associated with main- appears to reduce the prevalence of urinary incontinence in late tained or increased treatment effect, but this hypothesis needs pregnancy (34 weeks or more) and early postpartum (less than 12 further testing. weeks). Fifteen studies involving 6181 women (3040 PFMT, 3141 There is a need for at least one large, pragmatic, well conducted, controls) contributed to the analysis. Based on the trial reports, and explicitly reported randomised trial, comparing PFMT with pregnant women without prior urinary incontinence who were a control, to investigate the longer-term clinical effectiveness of randomised to intensive antenatal PFMT were less likely than PFMT. Also, studies investigating different pelvic floor muscle train- women randomised to no PFMT or usual antenatal care to report ing regimens are required to establish the optimum method of urinary incontinence in late pregnancy (about 56% less; RR 0.44, delivering and undertaking this intervention. 95% CI 0.30–0.65) and up to six months postpartum (about 30% In conclusion, pelvic floor exercises are beneficial and have no less; RR 0.71, 95% CI 0.52–0.97) [38–40]. significant adverse effects. Substantial and durable improvements Postnatal women with persistent urinary incontinence three in continence can be achieved, when the patient is appropriately months after delivery and who received PFMT were less likely than selected and the exercises are adequately performed. women who did not receive treatment (about 20% less; RR 0.79, 95% CI 0.70–0.90) to report urinary incontinence 12 months after Contributors delivery. The greatest treatment effect was seen in the trial with the most intensive, supervised strengthening PFMT programme (with Natalia Price reviewed the evidence and wrote the paper, the addition of weekly electrical stimulation). Based on the trials to Rehana Dawood reviewed the evidence and co-wrote the paper, date, the most beneficial population approach for postnatal PFMT and Simon R. Jackson edited the paper. appears to be to offer an individually taught strengthening PFMT programme to women potentially at greater risk of postnatal incon- Competing interests tinence, such as after a forceps delivery or vaginal delivery of a large baby. However, it seems that a PFMT programme of sufficient dose The authors have no competing interests to declare and were might be important both for women at potentially increased risk not in receipt of any funding to undertake this review. of postnatal incontinence and in a population-based approach to prevention of postnatal incontinence [38–40]. Provenance

Commissioned and externally peer reviewed. 11. Conclusions

References Overall, there is evidence for the widespread recommendation for use of pelvic floor muscle training as a first-line conserva- [1] Hunskaar S, Burgio K, Diokno AC, et al. Epidemiology and natural history of tive management programme for women with stress, urge or urinary incontinence. In: Abrams P, Cardozo L, Khoury S, editors. Incontinence. mixed urinary incontinence. A trial of supervised PFMT of at least 2nd ed. Health Publication; 2002. [2] Temml C, Haidinger G, Schmidbauer J. Urinary incontinence in both sexes: three months’ duration should be offered as first-line treatment to prevalence rates and impact on quality of life and sexual life. Neurourol Urodyn women with stress or mixed urinary incontinence. A pelvic floor 2001;19:259–71. muscle training programme should comprise at least eight con- [3] Mantle J. Physiotherapy for incontinence. In: Cardoza L, Staskin D, editors. Text- tractions performed three times per day. If pelvic floor muscle book of female urology and urogynaecology. London: Isis Medical Media Ltd; 2001. p. 351–8. training is beneficial, the exercise programme should be main- [4] National Institute for Health Clinical Excellence. Urinary incontinence: the tained. management of urinary incontinence in women. Clinical guideline 40. London: There have been many publications on the benefits of PFMT NICE; 2006. [5] Haslem J. Therapeutic management of incontinence and pelvic pain. 2nd ed. in urinary stress incontinence, although the evidence for its use London: Springer; 2007. in urge incontinence is more recent. There are no long-lasting or [6] Kegel AH. Progressive resistance exercise in the functional restoration of the debilitating adverse effects of pelvic floor muscle training. For those perineal muscles. Am J Obstet Gynaecol 1948;56:238–49. [7] Haylen, et al. An International Urogynecological Association cases where it does not succeed, there are other alternative man- (IUGA)/International Continence Society (ICS) joint report on the terminology agement options. PFMT is cost effective, uses fewer resources than a for female pelvic floor dysfunction. Int Urogynecol J 2010;21:5–26.

[8] Hannestad YS, Rortveit G, Sandvik H, Hunskaar S. A community base epidemi- [26] Miller JM, Ashton-Miller JA, DeLancey JOL. A pelvic muscle precontraction can ological survery of the female urinary incontinence: the Norwegian EPICONT reduce cough-related urine loss in selected women with mild SUI. J Am Geriatr study. J Clin Epidemiol 2000;53:1150–7. Soc 1998;46:870–4. [9] Rud T. Urethral pressure profile in continent women from childhood to old age. [27] Bo K, Talseth T, Holme I. Single blind, randomized controlled trial of pelvic Acta Obstet Gynaecol Scand 1980;59:331–5 [Pubmed 7192471]. floor exercises, electrical stimulation, vaginal cones and no treatment in man- [10] Enhorning G. Simultaneous recording of intravesical and intraurethral pres- agement of genuine stress incontinence in women. BMJ 1999;318:487–93. sure. Acta Obstet Gynaecol Scand Suppl 1961;276:1–69. [28] Herbison GP, Arnold EP. Sacral neuromodulation with implanted devices for [11] Howard D, Delancey JOL, Tunn R, Aston Miller JA. Racial Differences in the urinary storage and voiding dysfunction in adults. Cochrane Database Syst Rev structure and function of the stress continence mechanism. Obstet Gynaecol 2009. 2000;95(5):713–7. [29] Peters KM, Carrico DJ, Perez-Marrero RA, et al. Randomized trial of per- [12] Wilson PD, Herbison RM, Herbison JP. Obstetric practice, and the prevalence of cutaneous tibial nerve stimulation versus sham efficacy in the treatment urinary incontinence three months after delivery. BJOG 1996;103:154–61. of overactive bladder syndrome: results from the SUmit trial. J Urol [13] van Brummen HJ, Bruinse HW, van de Pol G, Heintz APM, van der Vaart CH. 2010;183:1438–43. The effect of vaginal and caesarean delivery on lower urinary tract symptoms: [30] Peters KM, MacDiarmid SA, Wooldridge LS, et al. Randomized trial of percu- what makes the difference? Int Urogyn J 2007;l18(3):133–9. taneous tibial nerve stimulation versus extended-release tolterodine: results [14] Fantl JA, Cardozo L, McClish DK. Estrogen therapy in the management from the overactive bladder innovative therapy trial. J Urol 2009;182:1055–61. of urinary incontinence in postmenopausal women: a meta analysis. First [31] Slack A, Hill A, Jackson S. Is there a role for a specialist physiotherapist report of the hormones and urogenital therapy committee. Obstet Gynecol in the multi-disciplinary management of women with stress incontinence 1994;83(1):12–8. referred from primary care to a specialist continence clinic? J Obstet Gynaecol [15] Burgio KL, Goode PS, Locher JL, et al. Behavioral training with and without 2008;28(4):410–2. biofeedback in the treatment of urge incontinence in older women: a random- [32] Gilling PJ, Wilson LC, Westenberg AM, et al. A double blind randomized ized controlled trial. JAMA 2002;288:2293–9. controlled trial of electromagnetic stimulation of the pelvic floor vs. sham [16] Cody JD, Richardson K, Moehrer B, Hextall A, Glazener CMA. Oestrogen therapy therapy in the treatment of women with stress urinary incontinence. BJU Int for urinary incontinence in post-menopausal women. Cochrane Database Syst 2009;103(10):1386–90. Rev 2009;(4). Art. No. CD001405. [33] Ishiko O, Hirai K, Sumi T, Tatsuta I, Ogita S. Hormone replacement therapy plus [17] Bo K. Pelvic floor muscle training is effective in treatment of female stress pelvic floor exercise for post menopausal stress incontinence: a randomised urinary incontinence, but how does it work? Int Urogynecol J 2004;15:76–84. controlled trial. J Reprod Med 2001;46(3):213–20. [18] Bo K, Lilleas F, Talseth T, Hedland H. Dynamic MRI of the pelvic floor muscles [34] Cammu H, Van Nylen M, Amy JJ. A 10 year follow up after Kegel’s in an upright sitting position. Neurourol Urodyn 2001;20:167–74. pelvic floor muscle exercises for genuine stress incontinence. Br J Urol Int [19] Bo K, Sherburn M. Evaluation of female pelvic-floor muscle function and 2001;85(6):655–8. strength. Physical Therapy 2005;85:269–82. [35] Godec C, Cass AS, Ayala GF. Bladder inhibition with functional electrical stim- [20] Dumoulin C, Hay Smith J. Pelvic floor muscle treatment versus no treatment, ulation. Urology 1975;6(6):663–6. or inactive control treatments, for urinary incontinence in women. Cochrane [36] de Groat WC. A neurologic basis for the overactive bladder. Urology 1997;50(6A Database Syst Rev 2010;(1). Art. No. CD005654. Suppl.):36–52. [21] Peschers U, Vodusek D, Fanger G, Schaer G, Delancey J, Schussler B. Pelvic [37] Morrison JFB. The excitability of the micturition reflex. Scand J Urol Nephrol muscle activity in nulliparous volunteers. Neurourol Urodyn 2001;20:269–75. 1995;29(Suppl. 175):21–5. [22] Abrams P, Andersson KE, Birder L, et al. Fourth international consultation on [38] Reilly ET, Freeman RM, Waterfield MR, Waterfield AE, Steggles P, Pedlar F. Pre- incontinence recommendations of the international scientific committee: eval- vention of postpartum stress incontinence in primigravidae with increased uation and treatment of urinary incontinence, pelvic organ prolapse, and fecal bladder neck mobility: a randomised controlled trial of antenatal pelvic floor incontinence. Neurourol Urodyn 2010;29:213–40. exercises. BJOG 2002;109(1):68–76. [23] Balmforth J, Mantle J, Ashton-Miller JA, Bidemead J, Cardozo L. A prospective [39] King JK, Freeman RM. Is antenatal bladder neck mobility a risk factor for post- observational trial of pelvic floor muscle training for female stress urinary partum stress incontinence? BJOG 1998;105:1300–7. incontinence. BJU Int 2006;98(4):811–7. [40] Hay-Smith J, Mørkved S, Fairbrother KA, Herbison GP. Pelvic floor muscle [24] Bo K, Bergman B, Morkved S, van Kampen M. Evidence based physical therapy training for prevention and treatment of urinary and faecal incontinence in for the pelvic floor. Edinburgh: Harrison, H; 2007. antenatal and postnatal women. Cochrane Database Syst Rev 2008;(4). Art. No. [25] Miller JM, Perucchini D, Cardichi LT, Delancey JOL, Ashton-Miller J. Pelvic floor CD007471. muscle contraction during a cough and decreased vesical neck mobility. Obstet Gynecol 2001;97:255–60.

Please cite this article in press as: Price N, et al. Pelvic ﬂoor exercise for urinary incontinence: A systematic literature review. Maturitas (2010), doi:10.1016/j.maturitas.2010.08.004